Journal versus clock tables: Opposing requirements?

Journal versus clock tables: Opposing requirements?

[Tommy Kelly]

I'm trying to get org-mode to provide me with two things, but haven't
found a way to do it.

1. First, I want to be able to use it like a daily engineering or
science journal, logging notes as they occur, in pretty much linear
fashion chronologically. Or, more to the point, I want to be able to
report and look at items as they occurred, in pretty much linear
fashion chronologically. Essentially I want to be able to report on
activity by time of occurrence, not topic.

2. But second, I want to see clock tables covering a period of time,
which groups related items together regardless of when (within the
given period) they happened. Essentially I want to be able to report
on actrivity by topic, not time of occurrence.

I'm using some of Bernt Hansen's excellent setup, but it still isn't
getting me quite where I want to be.

I'll note also that the agenda's log mode doesn't really give me point
1. It simply lists the *headlines* which have a clock entry or
timestamp at a given time. I want to see my entire journal -- a la a
blog. (*Ideally* I'd like to be able to control the depth to which
that entire journal output went to, but seeing the whole shebang would
be a good start.)

Anyone have any ideas how to do this.

thanks,
Tommy

Re: Journal versus clock tables: Opposing requirements?

[Bernt Hansen]

Tommy Kelly <tommy.kelly@verilab.com> writes:

> I'm trying to get org-mode to provide me with two things, but haven't
> found a way to do it.
>
> 1. First, I want to be able to use it like a daily engineering or
> science journal, logging notes as they occur, in pretty much linear
> fashion chronologically. Or, more to the point, I want to be able to
> report and look at items as they occurred, in pretty much linear
> fashion chronologically. Essentially I want to be able to report on
> activity by time of occurrence, not topic.
>
> 2. But second, I want to see clock tables covering a period of time,
> which groups related items together regardless of when (within the
> given period) they happened. Essentially I want to be able to report
> on actrivity by topic, not time of occurrence.
>
> I'm using some of Bernt Hansen's excellent setup, but it still isn't
> getting me quite where I want to be.
>
> I'll note also that the agenda's log mode doesn't really give me point
> 1. It simply lists the *headlines* which have a clock entry or
> timestamp at a given time. I want to see my entire journal -- a la a
> blog. (*Ideally* I'd like to be able to control the depth to which
> that entire journal output went to, but seeing the whole shebang would
> be a good start.)
>
> Anyone have any ideas how to do this.

Hi Tommy,

For item 1) can you use the display of inactive timestamps to get part
of the information you want in the agenda and then visit the items with
either follow mode (F) or manually visit each item with SPC to get more
detail?

If you create an inactive timestamp for every new note you log you can
display that timestamp with [ or ] in the agenda.  This is what I do.

For item 2) I would use agenda clock reports R while displaying the
agenda time frame you are interested in.  You can limit the agenda to
certain tags and use C-u R to limit the clock report to only those tags.

HTH,
Bernt

Re: Journal versus clock tables: Opposing requirements?

[Tommy Kelly]

Bernt wrote:
> For item 1) can you use the display of inactive timestamps to get part
> of the information you want in the agenda and then visit the items with
> either follow mode (F) or manually visit each item with SPC to get more
> detail?

Thanks. That's pretty much exactly my workaround now. So I enter data
all over the place in my file, so as to preserve position with respect
to headings (so my clock table is correct). Therefore the only way to
get the journal-style output seems to be as you suggest.

The problem is, at the end of the week, when I want to output a report
of what I did, it's a fairly manual task. It's true that even with a
chronological report I wouldn't necessarily leave the chronological
output as-is, with no editing or grouping. But it will be *much*
easier to get what I want if I can start with a simple linear-time
report of everything, than if I have to work my way through the weekly
agenda in follow mode.

Tommy

Re: Journal versus clock tables: Opposing requirements?

[Bernt Hansen]

Tommy Kelly <tommy.kelly@verilab.com> writes:

> Bernt wrote:
>> For item 1) can you use the display of inactive timestamps to get part
>> of the information you want in the agenda and then visit the items with
>> either follow mode (F) or manually visit each item with SPC to get more
>> detail?
>
> Thanks. That's pretty much exactly my workaround now. So I enter data
> all over the place in my file, so as to preserve position with respect
> to headings (so my clock table is correct). Therefore the only way to
> get the journal-style output seems to be as you suggest.
>
> The problem is, at the end of the week, when I want to output a report
> of what I did, it's a fairly manual task. It's true that even with a
> chronological report I wouldn't necessarily leave the chronological
> output as-is, with no editing or grouping. But it will be *much*
> easier to get what I want if I can start with a simple linear-time
> report of everything, than if I have to work my way through the weekly
> agenda in follow mode.

If you can manually create your report (even if it is tedious) then it
should be possible to automate most of it.  This is Emacs after all -
you can program it to do whatever you need it to do.

It should be possible to write code that walks your agenda, visits the
tasks, and copies and pastes the details to a temporary org buffer/file
just for your chronological report.

I have no idea what that code would look like though... it depends on
how often you require this weekly report if writing that code would be
worth it or not.

Regards,
Bernt

Re: Journal versus clock tables: Opposing requirements?

[Tommy Kelly]

> ... it depends on how often you require this weekly report...

Ermm, weekly :-)

> It should be possible to write code that walks your agenda, visits the
> tasks, and copies and pastes the details to a temporary org buffer/file
> just for your chronological report.

Absolutely. But as I've been thinking about this, I'm realizing that
in fact a key input requirement (as opposed to my two output
requirements -- chronology plus clock tables) is ease of entry. As I
begin some new chunk of work, I don't want to have to hunt around for
the most appropriate heading to clock into and begin writing notes
under. As a result, what's happening is two things. First, I'm just
falling back on your single "Organization" catch-all task; second, I'm
not writing *any* notes.

With a chronological journal, there's no decision to be made. You just
start logging at the end (or start) of the journal. But maybe
something in the org-capture area is what I need. I've tried it before
and didn't get very far, but I'll have another look.

thanks,
Tommy

Re: Journal versus clock tables: Opposing requirements?

[Carsten Dominik]

On Nov 8, 2011, at 9:06 AM, Tommy Kelly wrote:

>> ... it depends on how often you require this weekly report...
> 
> Ermm, weekly :-)
> 
>> It should be possible to write code that walks your agenda, visits the
>> tasks, and copies and pastes the details to a temporary org buffer/file
>> just for your chronological report.
> 
> Absolutely. But as I've been thinking about this, I'm realizing that
> in fact a key input requirement (as opposed to my two output
> requirements -- chronology plus clock tables) is ease of entry. As I
> begin some new chunk of work, I don't want to have to hunt around for
> the most appropriate heading to clock into and begin writing notes
> under. As a result, what's happening is two things. First, I'm just
> falling back on your single "Organization" catch-all task; second, I'm
> not writing *any* notes.
> 
> With a chronological journal, there's no decision to be made. You just
> start logging at the end (or start) of the journal. But maybe
> something in the org-capture area is what I need. I've tried it before
> and didn't get very far, but I'll have another look.

Yes, capture you notes into a date tree and turn on the clock whenever
you start a new entry!

- Carsten

Re: Journal versus clock tables: Opposing requirements?

[Bernt Hansen]

Tommy Kelly <tommy.kelly@verilab.com> writes:

>> ... it depends on how often you require this weekly report...
>
> Ermm, weekly :-)
>
>> It should be possible to write code that walks your agenda, visits the
>> tasks, and copies and pastes the details to a temporary org buffer/file
>> just for your chronological report.
>
> Absolutely. But as I've been thinking about this, I'm realizing that
> in fact a key input requirement (as opposed to my two output
> requirements -- chronology plus clock tables) is ease of entry. As I
> begin some new chunk of work, I don't want to have to hunt around for
> the most appropriate heading to clock into and begin writing notes
> under. As a result, what's happening is two things. First, I'm just
> falling back on your single "Organization" catch-all task; second, I'm
> not writing *any* notes.
>
> With a chronological journal, there's no decision to be made. You just
> start logging at the end (or start) of the journal. But maybe
> something in the org-capture area is what I need. I've tried it before
> and didn't get very far, but I'll have another look.

I tend to use capture mode to refile.org (all level 1 tasks) and then
refile them later to the appropriate place.  I clock time on the task
when I create it and sometimes I end the capture task with C-c C-c
(which stops the clock) and immediately switch back to it with F9-SPC in
my setup.  Sometime later I'll refile this item to the right place in
the tree.  I don't do that now - when I'm supposed to be working on it
but later when I have 2 minutes to spare :)

When I start something new it's C-M-r (or C-c r or whatever your capture
binding is) t or n (for Todo or Note) then start entering data in the
capture buffer that is being clocked.  C-c C-c when done and possible
switch back to it immediately with F9-SPC.

I never hunt around for where a task should live when I capture it - I
used to do that with multiple capture task targets and it was just too
slow.

Regards,
Bernt

Re: Journal versus clock tables: Opposing requirements?

[Bernt Hansen]

Tommy Kelly <tommy.kelly@verilab.com> writes:

>> ... it depends on how often you require this weekly report...
>
> Ermm, weekly :-)
>

Yes :) but if it's only for the next 3 weeks it's probably not worth the
coding effort.  If it's weekly for the indefinite future it might be.

-Bernt

Re: Journal versus clock tables: Opposing requirements?

[Olaf Dietsche]

Tommy Kelly <tommy.kelly@verilab.com> writes:

> I'm trying to get org-mode to provide me with two things, but haven't
> found a way to do it.
>
> 1. First, I want to be able to use it like a daily engineering or
> science journal, logging notes as they occur, in pretty much linear
> fashion chronologically. Or, more to the point, I want to be able to
> report and look at items as they occurred, in pretty much linear
> fashion chronologically. Essentially I want to be able to report on
> activity by time of occurrence, not topic.
>
> 2. But second, I want to see clock tables covering a period of time,
> which groups related items together regardless of when (within the
> given period) they happened. Essentially I want to be able to report
> on actrivity by topic, not time of occurrence.
>
> I'm using some of Bernt Hansen's excellent setup, but it still isn't
> getting me quite where I want to be.
>
> I'll note also that the agenda's log mode doesn't really give me point
> 1. It simply lists the *headlines* which have a clock entry or
> timestamp at a given time. I want to see my entire journal -- a la a
> blog. (*Ideally* I'd like to be able to control the depth to which
> that entire journal output went to, but seeing the whole shebang would
> be a good start.)

Maybe I misunderstand what you want to accomplish, but if you put your
journal into a separate file (e.g. journal.org), you could load it as
any other file into emacs and look at it. With org-cycle (C-u TAB) you
can fold everything and open selected entries (TAB on a single headline)
if you want.

You can create clock tables and select reported items by tags. So, if
you tag your journal entries, you can create clock tables made up of a
few entries only. See the org manual: "The clock table".

Regards, Olaf

Re: Journal versus clock tables: Opposing requirements?

[Tommy Kelly]

Olaf wrote:
> Maybe I misunderstand what you want to accomplish, but if you put your
> journal into a separate file (e.g. journal.org), ...

That's pretty much what I want. But if I do that I then have trouble
with getting sensible clock tables. For example, suppose I had:

*** Headline about some activity on Project A
CLOCK: [from]--[to] => duration
notes notes notes

*** Headline about some activity on Project B
CLOCK: [from]--[to] => duration
more notes notes note

*** Headline about some other activity on Project A
CLOCK: [from]--[to] => duration
yet more notes notes

Adding those as a chronological journal lets me get a report of
chronology, but it won't let me have a clock table with the times for
the two "on Project A" activities combined into one. Will it?

Of course I could shove the Project A activities under one single
higher-level headline, but that than violates the chronology side of
things (in the sense that I want to enter things into my journal in
chronological order, just as I would in a paper log book).

> You can create clock tables and select reported items by tags. So, if
> you tag your journal entries, you can create clock tables made up of a
> few entries only.

OK, that might be what I need then. I thought clock tables grouped
things by headings, not by tags. I'll have a look at the manual.

thanks,
Tommy

Re: Journal versus clock tables: Opposing requirements?

[Tommy Kelly]

> OK, that might be what I need then. I thought clock tables grouped
> things by headings, not by tags. I'll have a look at the manual.

I'm trying the tagging thing within clock tables, but I can't get it
working at all. I've attached a tag to a single headline, and checked
that I've got that right by using "C-c a m". Then I added a :tags item
to my clock table block but it seems to have no effect. Is this valid:

#+BEGIN: clocktable :maxlevel 2 :scope file :block thisweek :step week
:indent :tags test_tag
#+END:

I've also tried :tags 'test_tag', :tags '+test_tag' and a bunch of
other things, but nothing seems to do anything. What I was expecting
was that my clock table, currently filled with lots of items, would be
reduced to looking only at the single headline that I've tagged. But
it's not -- it just stays as it was before.

What am I doing wrong?
thx,
Tommy

Re: Journal versus clock tables: Opposing requirements?

[Olaf Dietsche]

[-- Attachment #1: Type: text/plain, Size: 1090 bytes --]

Tommy Kelly <tommy.kelly@verilab.com> writes:

>> OK, that might be what I need then. I thought clock tables grouped
>> things by headings, not by tags. I'll have a look at the manual.
>
> I'm trying the tagging thing within clock tables, but I can't get it
> working at all. I've attached a tag to a single headline, and checked
> that I've got that right by using "C-c a m". Then I added a :tags item
> to my clock table block but it seems to have no effect. Is this valid:
>
> #+BEGIN: clocktable :maxlevel 2 :scope file :block thisweek :step week
> :indent :tags test_tag
> #+END:
>
> I've also tried :tags 'test_tag', :tags '+test_tag' and a bunch of
> other things, but nothing seems to do anything. What I was expecting
> was that my clock table, currently filled with lots of items, would be
> reduced to looking only at the single headline that I've tagged. But
> it's not -- it just stays as it was before.

I haven't tried this myself, just looking at the manual. But playing
around, it seems you need double quotes around your tags match.

Here's a simple example for a start:


[-- Attachment #2: Small example --]
[-- Type: text/plain, Size: 1139 bytes --]

Tagged clock tables  -*- mode: org; fill-column: 78 -*-

* Shopping							    :errands:
  :CLOCK:
  CLOCK: [2011-11-07 Mo 09:45]--[2011-11-07 Mo 09:50] =>  0:05
  :END:
* Cleaning							      :house:
  :CLOCK:
  CLOCK: [2011-11-07 Mo 08:50]--[2011-11-07 Mo 09:05] =>  0:15
  :END:
* Kids								   :children:
  :CLOCK:
  CLOCK: [2011-11-07 Mo 08:30]--[2011-11-07 Mo 08:50] =>  0:20
  :END:
* Homework							   :children:
  :CLOCK:
  CLOCK: [2011-11-07 Mo 13:40]--[2011-11-07 Mo 14:05] =>  0:25
  :END:
* Exercises							     :health:
  :CLOCK:
  CLOCK: [2011-11-07 Mo 07:30]--[2011-11-07 Mo 07:50] =>  0:20
  :END:

#+BEGIN: clocktable :maxlevel 2 :scope file :tags "health"
Clock summary at [2011-11-08 Di 09:57]

| Headline     | Time   |
|--------------+--------|
| *Total time* | *0:20* |
|--------------+--------|
| Exercises    | 0:20   |
#+END: clocktable

#+BEGIN: clocktable :maxlevel 2 :scope file :tags "children"
Clock summary at [2011-11-08 Di 09:58]

| Headline     |   Time |
|--------------+--------|
| *Total time* | *0:45* |
|--------------+--------|
| Kids         |   0:20 |
| Homework     |   0:25 |
#+END: clocktable

[-- Attachment #3: Type: text/plain, Size: 15 bytes --]


Regards, Olaf

Re: Journal versus clock tables: Opposing requirements?

[Carsten Dominik]

On Nov 8, 2011, at 10:04 AM, Olaf Dietsche wrote:

> Tommy Kelly <tommy.kelly@verilab.com> writes:
> 
>>> OK, that might be what I need then. I thought clock tables grouped
>>> things by headings, not by tags. I'll have a look at the manual.
>> 
>> I'm trying the tagging thing within clock tables, but I can't get it
>> working at all. I've attached a tag to a single headline, and checked
>> that I've got that right by using "C-c a m". Then I added a :tags item
>> to my clock table block but it seems to have no effect. Is this valid:
>> 
>> #+BEGIN: clocktable :maxlevel 2 :scope file :block thisweek :step week
>> :indent :tags test_tag
>> #+END:
>> 
>> I've also tried :tags 'test_tag', :tags '+test_tag' and a bunch of
>> other things, but nothing seems to do anything. What I was expecting
>> was that my clock table, currently filled with lots of items, would be
>> reduced to looking only at the single headline that I've tagged. But
>> it's not -- it just stays as it was before.
> 
> I haven't tried this myself, just looking at the manual. But playing
> around, it seems you need double quotes around your tags match.

Yes.  The clock table parameters are read as lisp syntax, and the
match is a string.

HTH

- Carsten

> 
> Here's a simple example for a start:
> 
> Tagged clock tables  -*- mode: org; fill-column: 78 -*-
> 
> * Shopping							    :errands:
>  :CLOCK:
>  CLOCK: [2011-11-07 Mo 09:45]--[2011-11-07 Mo 09:50] =>  0:05
>  :END:
> * Cleaning							      :house:
>  :CLOCK:
>  CLOCK: [2011-11-07 Mo 08:50]--[2011-11-07 Mo 09:05] =>  0:15
>  :END:
> * Kids								   :children:
>  :CLOCK:
>  CLOCK: [2011-11-07 Mo 08:30]--[2011-11-07 Mo 08:50] =>  0:20
>  :END:
> * Homework							   :children:
>  :CLOCK:
>  CLOCK: [2011-11-07 Mo 13:40]--[2011-11-07 Mo 14:05] =>  0:25
>  :END:
> * Exercises							     :health:
>  :CLOCK:
>  CLOCK: [2011-11-07 Mo 07:30]--[2011-11-07 Mo 07:50] =>  0:20
>  :END:
> 
> #+BEGIN: clocktable :maxlevel 2 :scope file :tags "health"
> Clock summary at [2011-11-08 Di 09:57]
> 
> | Headline     | Time   |
> |--------------+--------|
> | *Total time* | *0:20* |
> |--------------+--------|
> | Exercises    | 0:20   |
> #+END: clocktable
> 
> #+BEGIN: clocktable :maxlevel 2 :scope file :tags "children"
> Clock summary at [2011-11-08 Di 09:58]
> 
> | Headline     |   Time |
> |--------------+--------|
> | *Total time* | *0:45* |
> |--------------+--------|
> | Kids         |   0:20 |
> | Homework     |   0:25 |
> #+END: clocktable
> 
> Regards, Olaf

- Carsten

Re: Journal versus clock tables: Opposing requirements?

[Tommy Kelly]

> I haven't tried this myself, just looking at the manual. But playing
> around, it seems you need double quotes around your tags match.

Ah OK, that works. But it turns out there was a second problem and it
may be a bug. It looks like any of the clocktable options after
:indent get ignored. So this works:

#+BEGIN: clocktable :maxlevel 2 :scope file :block thisweek :step week
:tags "test_tag" :indent
#+END:

But this doesn't (i.e. the :tags option is ignored)

#+BEGIN: clocktable :maxlevel 2 :scope file :block thisweek :step week
:indent :tags "test_tag"
#+END:

Tommy

Re: Journal versus clock tables: Opposing requirements?

[Carsten Dominik]

On Nov 8, 2011, at 10:55 AM, Tommy Kelly wrote:

>> I haven't tried this myself, just looking at the manual. But playing
>> around, it seems you need double quotes around your tags match.
> 
> Ah OK, that works. But it turns out there was a second problem and it
> may be a bug. It looks like any of the clocktable options after
> :indent get ignored. So this works:
> 
> #+BEGIN: clocktable :maxlevel 2 :scope file :block thisweek :step week
> :tags "test_tag" :indent
> #+END:
> 
> But this doesn't (i.e. the :tags option is ignored)
> 
> #+BEGIN: clocktable :maxlevel 2 :scope file :block thisweek :step week
> :indent :tags "test_tag"


You need 

   :indent t

This is a property list, each key needs a value.

- Carsten

Re: Journal versus clock tables: Opposing requirements?

[Sebastien Vauban]

Hi Carsten,

Carsten Dominik wrote:
> On Nov 8, 2011, at 10:55 AM, Tommy Kelly wrote:
>
>>> I haven't tried this myself, just looking at the manual. But playing
>>> around, it seems you need double quotes around your tags match.
>> 
>> Ah OK, that works. But it turns out there was a second problem and it
>> may be a bug. It looks like any of the clocktable options after
>> :indent get ignored. So this works:
>> 
>> #+BEGIN: clocktable :maxlevel 2 :scope file :block thisweek :step week
>> :tags "test_tag" :indent
>> #+END:
>> 
>> But this doesn't (i.e. the :tags option is ignored)
>> 
>> #+BEGIN: clocktable :maxlevel 2 :scope file :block thisweek :step week
>> :indent :tags "test_tag"
>
> You need
>
>    :indent t
>
> This is a property list, each key needs a value.

This makes sense. But why did his first example work, then?  There is no `t'
either.

Best regards,
  Seb

-- 
Sebastien Vauban

Re: Journal versus clock tables: Opposing requirements?

[Carsten Dominik]

On Nov 8, 2011, at 11:05 AM, Sebastien Vauban wrote:

> Hi Carsten,
> 
> Carsten Dominik wrote:
>> On Nov 8, 2011, at 10:55 AM, Tommy Kelly wrote:
>> 
>>>> I haven't tried this myself, just looking at the manual. But playing
>>>> around, it seems you need double quotes around your tags match.
>>> 
>>> Ah OK, that works. But it turns out there was a second problem and it
>>> may be a bug. It looks like any of the clocktable options after
>>> :indent get ignored. So this works:
>>> 
>>> #+BEGIN: clocktable :maxlevel 2 :scope file :block thisweek :step week
>>> :tags "test_tag" :indent
>>> #+END:
>>> 
>>> But this doesn't (i.e. the :tags option is ignored)
>>> 
>>> #+BEGIN: clocktable :maxlevel 2 :scope file :block thisweek :step week
>>> :indent :tags "test_tag"
>> 
>> You need
>> 
>>   :indent t
>> 
>> This is a property list, each key needs a value.
> 
> This makes sense. But why did his first example work, then?  There is no `t'
> either.

Probably, :indent was perceived by the code as nil, but at least
it did not swallow the :tags key.....

- Carsten

Re: Journal versus clock tables: Opposing requirements?

[Sebastien Vauban]

Hi Carsten,

Carsten Dominik wrote:
> On Nov 8, 2011, at 11:05 AM, Sebastien Vauban wrote:
>
>> Hi Carsten,
>> 
>> Carsten Dominik wrote:
>>> On Nov 8, 2011, at 10:55 AM, Tommy Kelly wrote:
>>> 
>>>>> I haven't tried this myself, just looking at the manual. But playing
>>>>> around, it seems you need double quotes around your tags match.
>>>> 
>>>> Ah OK, that works. But it turns out there was a second problem and it
>>>> may be a bug. It looks like any of the clocktable options after
>>>> :indent get ignored. So this works:
>>>> 
>>>> #+BEGIN: clocktable :maxlevel 2 :scope file :block thisweek :step week
>>>> :tags "test_tag" :indent
>>>> #+END:
>>>> 
>>>> But this doesn't (i.e. the :tags option is ignored)
>>>> 
>>>> #+BEGIN: clocktable :maxlevel 2 :scope file :block thisweek :step week
>>>> :indent :tags "test_tag"
>>> 
>>> You need
>>> 
>>>   :indent t
>>> 
>>> This is a property list, each key needs a value.
>> 
>> This makes sense. But why did his first example work, then?  There is no `t'
>> either.
>
> Probably, :indent was perceived by the code as nil, but at least
> it did not swallow the :tags key.....

OK, clear.

Thanks,
  Seb

-- 
Sebastien Vauban

Re: Journal versus clock tables: Opposing requirements?

[Tommy Kelly]

> Probably, :indent was perceived by the code as nil, but at least
> it did not swallow the :tags key.....

It wasn't. I hadn't realized about the need for a value to the
property, but if it's omitted then it looks like :indent's value is
perceived as true, not nil (which is why I got on so long not
realizing the value was required).

Tommy

Re: Journal versus clock tables: Opposing requirements?

[Eric S Fraga]

Olaf Dietsche <olaf+list.orgmode@olafdietsche.de> writes:

> Tommy Kelly <tommy.kelly@verilab.com> writes:
>
>>> OK, that might be what I need then. I thought clock tables grouped
>>> things by headings, not by tags. I'll have a look at the manual.

[...]

> #+BEGIN: clocktable :maxlevel 2 :scope file :tags "health"
> Clock summary at [2011-11-08 Di 09:57]
>
> | Headline     | Time   |
> |--------------+--------|
> | *Total time* | *0:20* |
> |--------------+--------|
> | Exercises    | 0:20   |
> #+END: clocktable

I've lost the rest of this thread and I don't actually know who to
attribute the original statement above...

In any case, I would like to have entries in the clock report by tag as
opposed to headline (as alluded to above).  Is this actually possible?
My logging approach is to have descriptive headings (the specific task
undertaken) with tags indicating the type of activity (teaching,
research, admin, personal).  I need a breakdown of my time spent in each
category in one table. 

I realise I can do N tables, one for each tag, but this is not
particularly useful for me.  It'll do but if there's a nicer solution,
I'll take it!

The documentation refers to a :formatter entry, but I am not sure
whether this is relevant as it says that this is for formatting an
entry.  My problem is that I want tag based entries, not how they are
formatted?

Apologies if I've missed something obvious in the manual!

Thanks,
eric

-- 
Eric S Fraga (GnuPG: 0xC89193D8FFFCF67D)

Re: Journal versus clock tables: Opposing requirements?

[Olaf Dietsche]

Eric S Fraga <e.fraga@ucl.ac.uk> writes:

> Olaf Dietsche <olaf+list.orgmode@olafdietsche.de> writes:
>
>> Tommy Kelly <tommy.kelly@verilab.com> writes:
>>
>>>> OK, that might be what I need then. I thought clock tables grouped
>>>> things by headings, not by tags. I'll have a look at the manual.
>
> [...]
>
>> #+BEGIN: clocktable :maxlevel 2 :scope file :tags "health"
>> Clock summary at [2011-11-08 Di 09:57]
>>
>> | Headline     | Time   |
>> |--------------+--------|
>> | *Total time* | *0:20* |
>> |--------------+--------|
>> | Exercises    | 0:20   |
>> #+END: clocktable
>
> I've lost the rest of this thread and I don't actually know who to
> attribute the original statement above...

Do you mean
<http://www.mail-archive.com/emacs-orgmode@gnu.org/msg48633.html>?

> In any case, I would like to have entries in the clock report by tag as
> opposed to headline (as alluded to above).  Is this actually possible?
> My logging approach is to have descriptive headings (the specific task
> undertaken) with tags indicating the type of activity (teaching,
> research, admin, personal).  I need a breakdown of my time spent in each
> category in one table. 
>
> I realise I can do N tables, one for each tag, but this is not
> particularly useful for me.  It'll do but if there's a nicer solution,
> I'll take it!
>
> The documentation refers to a :formatter entry, but I am not sure
> whether this is relevant as it says that this is for formatting an
> entry.  My problem is that I want tag based entries, not how they are
> formatted?
>
> Apologies if I've missed something obvious in the manual!

Regards, Olaf